Windows strategy generator — genetic programming writes the rules, you export the code.
$995/mo
Forex data and FX trading tools
Every card here raises the same question first: whose prices are these, and against whom are you trading them?
Last updated
There is no exchange, so there is no price list — and no single price
Everything about this page follows from one structural fact: spot foreign exchange is an over-the-counter market. There is no central venue, so there is no exchange fee schedule, no consolidated tape and no per-device licence. That is why a plan selling US equity quotes for tens of dollars a month will throw FX in for nothing, and why twenty-five of the cards here are data APIs that treat currencies as a rounding error on the bill.
The cost of that is precision. A quote for a pair is a quote from somebody's book, and two vendors will disagree about the same pair at the same instant by more than the difference is usually worth arguing about. In equities you can ask what the official close was. In FX the honest question is: whose close?
The three kinds of FX price on this page
Official reference rates. One number per currency per day, published by an institution, not dealable. The ECB Data Portal API serves the daily euro reference rates, yield curves and €STR over keyless SDMX across 214 dataflows; FRED carries FX among its macro series behind a free key; IMF Data, DBnomics and EconDB aggregate the same class of series from statistics offices and central banks. These are the right source for valuation, restatement and anything a third party has to be able to check.
Vendor consolidated quotes. A price built from a vendor's own sources and sold as a feed. Twelve Data covers FX alongside global equities and crypto over REST and WebSocket, billed in credits per minute; Tiingo, Finnhub, EODHD, Alpha Vantage and Financial Modeling Prep all do a version of the same thing. This is the layer where "which source" stops being answerable and starts being a vendor implementation detail.
Tick history. Tick Data sells spot FX by the symbol-year at 160 dollars, starting 1 May 2008. Kibot and FirstRate Data sell bought-once archives as zipped CSV. IQFeed streams it tick by tick into a local Windows socket. If a backtest depends on the spread rather than on the mid, this is the layer that answer lives in.
Picking between the three is not a quality judgement, it is a question about what the number is for. A reference rate cannot be traded and a vendor quote cannot be cited; asking a single source to do both jobs is how a revaluation ends up disagreeing with an accountant and a backtest ends up disagreeing with a broker statement. It is normal to use two of these at once, and it costs almost nothing to do so, because the free and near-free layer here is unusually good.
Where the platforms sit
Twelve charting platforms here carry FX, and for most of them it is an additional instrument list rather than a different product — TradingView, Koyfin, Optuma, StockCharts. The ones where FX is structurally central are the execution platforms: MultiCharts and NinjaTrader route orders to a broker and bring no data of their own, which in FX means the broker's prices are the prices — the same book you are dealing against is the book you are charting.
That is worth saying plainly, because it is the difference between FX and every other asset class in this catalogue. Elsewhere, the data source and the counterparty are separate companies with separate incentives. In retail FX they are frequently the same firm.
The cards here for a different reason
Sixteen backtesting frameworks list FX — NautilusTrader, Backtrader, vectorbt, QuantConnect and the rest. They consume prices rather than supply them, and the choice of source above is what determines whether their output means anything.
Four analysis libraries are here and none of them is a data tool. arch fits GARCH and the rest of the volatility-model family, which is the canonical FX application of that family; QuantLib builds the curves; TA-Lib and pandas-ta compute indicators on whatever series you hand them.
Two portfolio trackers appear, for unrelated reasons. Delta carries FX as part of a mixed stock-and-crypto book rather than as a trading instrument; TradeLog is here for how FX gains are taxed, which is a question about your jurisdiction rather than about the data. Six trade journals accept FX fills for the same reason journals accept anything: it is a record of what you did, not a view on what to do next. For the free layer under all of this, genuinely free financial data is the wider map.
All 72 of them
Showing 72 of 72
The API most people's first script talks to — free key, wide coverage, hard rate limits.
$49.99/moFree tierMCP
Windows portfolio backtester scripted in AFL, sold as a perpetual licence.
$189/mo
GARCH and the rest of the volatility-model family, plus the tests you need around them.
FreeOpen source
A single-instrument Python backtester — one OHLC series, one strategy, no live trading.
FreeOpen source
An event-driven Python backtester with 122 indicators, frozen since April 2023.
FreeOpen source
Seventy-odd REST endpoints covering equities, options and physical commodities.
—
Stock, fund and options screeners sold as a $0-$29.95 website membership.
$9.99/moFree tier
Free hosted Python backtesting with bundled minute data and broker execution.
Free
Point-and-click strategy generation on Windows, exported as code to eight platforms.
$1497 once
Per-file OPRA options history from the exchange, priced by dataset, date and symbol.
$50/mo
End-of-day futures history back to 1922 and eight ways to build a continuous contract.
$20/mo
Macro series from 93 statistics offices and central banks behind one open API.
€1500/yrFree tier
One valuation view over a mixed stock-and-crypto book, owned by a broker.
$6.99/moFree tier
Euro reference rates, yield curves and €STR over keyless SDMX — with vintages.
Free
5,170 macro series and container-shipping data behind one API and one subscription.
€24.99/moFree tier
Post-trade analytics and discipline tracking on one annual plan, with no free tier.
$197/yr
End-of-day and fundamentals for 60+ exchanges worldwide, at a hobbyist price.
$19.99/moFree tierMCP
Financial statements, ratios and filings for 70,000+ securities across 60+ exchanges.
$22/moFree tierMCP
Real-time US quotes and a live trade stream on a free key. OHLC candles are not free.
$49.99/moFree tierMCP
Real-time quotes, advanced filters and CSV export on top of the free Finviz screener.
$39.50/moFree tier
Historical US intraday and tick data as zipped CSV, bought once rather than rented.
$239.94/moFree tier
The Fed's macro database — 851,100 series, a free key and 120 requests a minute.
Free
The community continuation of ib_insync — same API, new maintainers, TWS still required.
FreeOpen source
Free, keyless SDMX access to IMF macro statistics — WEO included, forecasts and all.
Free
Tick-by-tick data into a local Windows socket, with exchange fees billed on top.
$20/mo
US intraday history back to 1998, bought once and kept, delivered as gzipped CSV.
$14/moFree tier
A browser research terminal for global equities, funds, macro and estimates.
$49/moFree tier
One strategy class for backtest and live, plus a built-in LLM agent runtime.
$24.50/moFree tier
Global financial news over REST, tagged to entities and scored for sentiment per entity.
$29/moFree tier
Full-tick US equities, options and futures from a direct exchange feed.
$29/moFree tierMCP
Java desktop charting whose Elliott Wave tooling is the reason anyone buys it.
$24/moFree tier
Windows charting, backtesting and order routing on EasyLanguage, sold outright or monthly.
$49/moFree tier
Rust core, Python API, one strategy that backtests and trades live unchanged.
Free tier onlyOpen source
Windows futures platform with C# strategy scripting and order-flow charting.
$99/moFree tier
Survivorship-bias-free end-of-day history with delisted stocks and index constituents.
$150/yr
Enterprise tick capture, storage and streaming analytics, sold only by quote.
—
One Python API, REST app and MCP server over thirty other people's data feeds.
$500/yrFree tierMCP
Windows technical-analysis desktop for analysts, with Optuma's own EOD data bundled in.
$810/yr
Short interest estimated daily from the lending pool, with cost to borrow beside it.
$49/moFree tierMCP
The DataFrame indicator library whose repo is gone and whose last release is a year old.
FreeOpen source
The open-source LEAN engine, plus a hosted cloud that runs it against real brokers.
$84/moFree tierMCP
The open-source derivatives pricing library banks actually use, reachable from Python.
FreeOpen source
Windows order-flow terminal for 60+ connections, with an open C# API and a real free tier.
$40/moFree tier
Zipline and Moonshot in Docker on your own hardware, wired to Interactive Brokers.
Free tier only
Forty years of global fundamentals over REST, an MCP server and Google Sheets.
$29/moFree tierMCP
Windows-native charts, order flow and DOM trading, priced as packages plus exchange fees.
$26/mo
Browser technical charting, custom scans and alerts, priced per service level.
$19.95/moFree tier
C#/.NET algo stack with 384 connectors; source is published but no longer open source.
$995/yrFree tier
Machine-searches for trading strategies, then tries to break them with robustness tests.
$1290 once
The C indicator library everything else wraps — and pip now ships the C part with it.
FreeOpen source
Historical intraday ticks from 1974, sold by the symbol-year and licensed perpetually.
$250/mo
End-of-day equity history back to 1962, plus crypto, forex and news, from $30 a month.
$30/moFree tier
Wash sales and Form 8949 rebuilt from raw broker trade history, not from the 1099-B.
$219/yr
Multi-asset trade journal and tracker with a free plan and a 194-name import list.
$19.95/moFree tier
Options-first trade journal - multi-leg spreads classified on import, replay to 250ms.
$29.95/mo
The category's oldest journal, 81 supported brokers, and a free tier that still works.
$29.95/moFree tier
The analytics-heavy journal: 150+ base charts across seven instrument types.
$19.99/moFree tier
Auto-synced trade journal with bar-by-bar replay, backtesting and AI session review.
$35/mo
Macro data for 196 countries with its own forecasts. No free key; the API starts at $199.
$29.99/moMCP
Analyst, bull, bear, risk and portfolio-manager agents argue over one ticker.
FreeOpen source
Charting across 50-plus markets, with Pine Script, server-side alerts and broker trading.
$12.95/moFree tier
Screeners, technical analysis and backtests for any MCP client. No TradingView account.
$9/moFree tierMCP
Automated technical analysis, real-time scanning and no-code backtesting in one web app.
$82/mo
Global stocks, forex and crypto over REST and WebSocket, billed in credits per minute.
$29/moFree tierMCP
US options tape, dark pool prints and congressional trades over one REST API.
$50/moFree tierMCP
Vectorised backtesting — thousands of parameter combinations in one NumPy pass.
Free
Paid vectorbt — native Rust simulators, streaming indicators and a built-in MCP server.
$25/moMCP
A local finance research agent with 74 MCP tools, nine backtest engines and 14 brokers.
FreeOpen sourceMCP
Windows portfolio backtesting in C# or drag-and-drop blocks, with a built-in MCP service.
$49.95/moMCP
Financials, estimates and live prices as worksheet functions in Excel and Google Sheets.
$60/yr
The Python library everyone starts with — free, unofficial, and not something to build on.
FreeOpen source
FAQ
Why is forex data so much cheaper than stock data?
Because there is no exchange to pay. Spot FX trades over the counter, between banks and brokers, so there is no central venue setting a fee schedule and no consolidated tape to license. What you get instead is a price from somebody's book, and different vendors will give you different numbers for the same pair at the same instant. Cheap data and ambiguous data are the same fact here, seen from two sides.
What is a central bank reference rate good for?
Anything that has to be defensible rather than tradable — valuing a portfolio at a date, restating a foreign revenue line, writing a rule that both sides of a contract can check. The ECB publishes a daily euro reference rate set at a fixed time; it is one number a day, it is not a price you can deal on, and that is exactly why it settles arguments.
Does it matter which vendor I take FX history from?
More than it does for listed equities. An exchange-traded stock has one official close; a currency pair has as many closes as there are sources, differing by the spread and by the snapshot time. For backtesting this is a real source of divergence between your results and someone else's, so record which source you used and at what timestamp convention, the same way you would record a data version.
Why are macro APIs listed as forex tools?
Because they carry the official rates, and for a lot of work those are the right ones. FRED, the ECB Data Portal, DBnomics, IMF Data and EconDB all publish FX series as part of a macro catalogue rather than as a trading product. If the job is a chart, a revaluation or a research note rather than an order, these are free, keyless or nearly so, and they cite their own origin.